Georges Canguilhem, What Is a Scientific Ideology?, 1969


What is a scientific ideology? This is a question that arises, or so it seems to me, in the practice of the history of science, and its answer may be of importance for the theory of that subject. Perhaps the first question to ask is what it is that the history of science claims to be the history of. An easy answer is that the history of science is the history of a certain cultural form called “science”. One must then specify precisely what criteria make it -possible to decide whether or not, at any given time, a particular practice or discipline merits the name science. And it is precisely a question of merit, for “science” is a kind of title, a dignity not to be bestowed lightly. Hence another question becomes inevitable: Should the history of science exclude or, on the contrary, should it tolerate or even include the history of the banishment of inauthentic knowledge from the realm of authentic science? I use the word banishment quite intentionally, for what is at stake is nothing less than the legal withdrawal of legitimately acquired privileges. We have long since ceased to believe as Voltaire believed, that superstitions and false beliefs were invented by cynical dervishes and foisted upon the innocent by ignorant nursemaids 1.

Obviously this is more than a question of historical method or technique concerning what can be learned about the past of science from documents and archives. It is really an epistemological problem concerning the way in which scientific knowledge is historically constituted. Professor Suchodolski has recently posed a similar question:

“If the whole history of science up to the present time were in fact the history of ‘antiscience’, that would no doubt prove that it could not have been otherwise and probably that it will not be otherwise in the future … The history of science as a history of truth cannot be written. It is an oxymoron.” 2

I shall have more to say about the concept of “antiscience”. In particular, I shall examine the extent to which it coincides with what might be meant by the word ideology.

The question of ideology arises, as I said, in connection with the practice of the history of science, although many practicing historians have never bothered to ask it. Surprisingly, those who have asked have been vague about the criteria by which ideology is defined. Few historians of mathematics, for instance, have looked at the magical or mystical properties of numbers and shapes as part of their subject. Historians of astronomy do pay some attention to astrology, despite the fact that Copernicus in 1543 exploded the specious foundations of “horoscopic science”. But they do so only because astronomers were indebted to astrologers for several centuries’ observation of the heavens. Many historians of chemistry are aware of the history of ·alchemy and regard alchemy as a “stage” in the inception of chemical science. Historians of human sciences such as psychology find the past of their subject more embarrassing. Two-thirds of Brett’s history of psychology is devoted to theories of the soul, consciousness, and life of the spirit, many of which predate the very term psychology and, a fortiori, the modern concept associated with it.


Is the notion of scientific ideology relevant? Is the term a suitable one to designate and properly delimit the whole range of discursive structures claiming to be theories, the whole variety of more or less consistent representations of interphenomenal relations, and the whole spectrum of more or less permanent structures in terms of which men have interpreted their everyday experience? In short, is it a useful way of denoting those pseudosciences whose falsity is revealed solely by the fact that a genuine science has been established to refute their claims?

There is no mystery about the reasons for the wide­spread use of the word ideology today. It stems from the vulgarization of Karl Marx’s thought. Ideology is an epistemological concept with a polemical function, applied to systems of representation that express themselves in the languages of politics, ethics, religion, and metaphysics. These languages claim to express things as they are, whereas in reality they are means of protecting and defending ·a situation, that is, a particular structure of the relations between men and things. Marx attacked ideology in the name of the science that he claimed to be instituting the science of men making their own history, though not necessarily the history they wished to be making.

Marx borrowed the term ideology from eighteenth-century French philosophy. Cabanis and Destutt de Tracy defined it as the science of the genesis of ideas. The idéologues, as their followers were called, proposed treating ideas as natural phenomena determined by the relation between man, a living, sensitive organism, and his natural environment. Positivists before the fact, the ideologues were nevertheless liberals, opponents of the theologians and metaphysicians of their day. At first these liberals were deceived by Napoleon’s political maneuvers; they believed him to be the heir of the French Revolution. But when they turned against him, Napoleon heaped scorn and irony upon their heads, and it was he who was responsible for distorting their public image 3. Ideology was denounced in the name of political realism (according to which laws were to be based on knowledge of the human heart and the lessons of history) as mere metaphysics, thought without content.

In the meaning that Marx gave to the term, he preserved the idea that ideology inverts the relation between knowledge and the thing known. Ideology, which initially denoted the natural science of man’s acquisition of ideas about reality, came to be a term applied to any system of ideas resulting from a situation in which men were prevented from understanding their true relation to reality. Ideology exists, according to Marx, wherever attention is diverted from its proper object.

Can the notion of scientific ideology be subsumed without distortion under the general notion of ideology in the Marxist sense? At first sight, the answer is no. In The German Ideology Marx draws a sharp contrast between political, legal, economic, and religious ideologies and economic science, by which he meant the economic science that he intended to institute. Science authenticates itself, he argued, by tearing the veil that is ideology’s only substance. Hence scientific ideology is a contradiction in terms. By definition, every ideology stands at a distance from reality; every ideology fails to touch the true object that it believes it is examining. Marx sets out to prove that, compared with the Marxist science of economics, all political and economic ideologies are determined by the class position of the bourgeois intellectual, who thinks that he is looking at the reflection of things themselves as in a mirror when in fact all he sees is an inverted image of man’s relation to other men and to nature. No ideology speaks the truth. Although some are less removed from reality than others, all are illusory 4. And by illusory he means not simply mistaken but also comforting: ideologies are reassuring fables, unconsciously complicit in a judgment determined by self-interest 5. In short, Marx holds that ideology performs a compensatory function. Bourgeois ideologies are reactions, symptoms of social conflict and class struggle, yet as theories they are wont to deny the concrete problems without which they would not exist.

But, someone will rightly object, is it not noteworthy that Marx never counts science among the ideologies discussed in The German Ideology? It is indeed. To be sure, in his critique of Feuerbach, Marx charges that the philosopher failed to understand that so-called pure science takes its aims and its means from commerce and industry, or, in other words, from man’s material activity. But does this imply that there is no difference in epistemological status between, say, liberal political economy, which for Marx is an ideological discourse, and such well-tested theories as electromagnetism or celestial mechanics? It is quite true that the development of astronomy in the seventeenth and eighteenth centuries depended on the fabrication of optical and chronometric instruments. In the eighteenth century, the determination of longitude on the high seas was a theoretical issue, but the theory in question drew upon the clockmaker’s art to develop a commercially valuable technology. Today, however, the celestial mechanics of Newton is being confirmed experimentally on a grand scale through space programs supported by technologies and economies informed by quite different ideologies. To say that the science of nature is not independent of the mode of production and exploitation of nature is not to say that the problems and methods of science are not autonomous; unlike economic or political theory, science is not thereby subordinated to the dominant ideology of the ruling class at a particular moment in the history of society. In his Critique of Political Economy Marx encountered what he termed a “difficulty”, namely, that art, though produced under specific social conditions, could maintain its value, even after those conditions had disappeared. Can Marxism refuse to Greek geometry what Marx granted to Greek art?

Yet even if scientific knowledge cannot be placed under the head of ideology, is there any reason why we cannot give a meaning to the concept of scientific ideology? In the category of ideology a distinction needs to be made between content and function. Marx explicitly states that ideologies will cease to exist when the class whose destiny is to abolish all classes has accomplished its dialectical mission. The function of ideology – to delude – will no longer exist. Of course Marx is assuming that his description of the pacified, classless society is correct. History continues, however, after this stage is reached, and one might even say that it just begins. It is now the history not of class struggle but of man’s relation to nature. A new question then arises. Can one predict the development of man’s new relation to nature? In other words, can one foresee a tranquil and orderly future for the history of science? Or will the production of new scientific knowledge in the future require, as it has required in the past, lucky discoveries that can be rationally exploited only after they have been made? In order to establish man’s new relation to nature, will not men have to go beyond what is already known and verified? If so, then scientific ideology would be both an obstruction to and a necessary precondition of progress. The history of science would need to include a history of scientific ideologies, explicitly recognized as such. Let me therefore try to show the usefulness of the concept.


Scientific ideology, unlike a political class ideology, is not false consciousness. Nor is it false science. The essence of false science is that it never encounters falsehood, never renounces anything, and never has to change its language. For a false science there is no prescientific state. The assertions of a false science can never be falsified. Hence false science has no history. By contrast, a scientific ideology does have a history. A scientific ideology comes to an end when the place that it occupied in the encyclopedia of knowledge is taken over by a discipline that operationally demonstrates the validity of its claim to scientific status, its “norms of scientificity”. At that point a certain form of nonscience is excluded from the domain of science. I say nonscience rather than use Suchodolski’s term “antiscience” simply in order to take note of the fact that in a scientific ideology there is an explicit ambition to be science, in imitation of some already constituted model of what science is. This is a crucial point. The existence of scientific ideologies implies the parallel and prior existence of scientific discourses. Hence it also presupposes that a distinction has already been made between science and religion.

Consider the case of atomism. Democritus, Epicurus, and Lucretius claimed scientific status for their physics and psychology. To the antiscience of religion they opposed the antireligion of science. Scientific ideology neglects the methodological requirements and operational possibilities of science in the realm of experience that it chooses to explore, but it is not ignorance and it does not scorn or repudiate the function of science. Hence scientific ideology is by no means the same thing as superstition, for ideology has its place, possibly usurped, in the realm of knowledge, not in the realm of religious belief. Nor is it superstition in the strict etymological sense. A superstition is a belief from an old religion that persists despite its prohibition by a new religion. Scientific ideology does indeed stand over (super-stare) a site that will eventually be occupied by science. But science is not merely overlain; it is pushed aside (deportare) by ideology. Therefore, when science eventually supplants ideology, it is not in the expected location. When chemistry and physics established scientific knowledge of the atom in the nineteenth century, the place of the atom was not the one assigned to it in atomistic ideology: the place of the indivisible. What science finds is not what ideology suggested looking for. The persistence of the word proves nothing when the context and methods differ as much as the technique of pulverization differs from the methods of modern atomic research. Indeed, what the ideology heralded as simple reveals in its scientific reality a hierarchy of complexities.

For another, I hope convincing, example of the way in which scientific ideologies are supplanted by science, consider the Mendelian theory of heredity. Most historians of biology believe that Maupertuis was the forerunner of modern genetics because in his Vénus physique he considered the mechanisms by which normal and abnormal traits are transmitted, he used the calculus of probabilities to decide whether the frequency of a particular abnormality within a particular family was or was not fortuitous, and he explained hybridization by assuming the existence of seminal atoms, hereditary elements that combined during copulation. But it is enough to compare the writings of Maupertuis and Mendel to see the magnitude of the ·gap between a science and the ideology that it replaces. The facts that Mendel studies are not those gleaned by a casual observer; they are obtained through systematic research. That research was dictated by the nature of Mendel’s problem, for which there is no precedent in the pre-Mendelian literature. Mendel invented the idea of a character, by which he meant not the elementary agent of hereditary transmission but the element of heredity itself. A Mendelian character could enter into combination with n other characters, and one could measure the frequency of its appearance in successive generations. Mendel was not interested in structure, fertilization, or development. For him, hybridization was not a way of establishing the constancy or inconstancy of a global type; it was a way of decomposing a type, an instrument of analysis, a tool for separating characters that made it necessary to work with large samples. Hence Mendel was interested in hybrids despite his repudiation of an age-old tradition of hybrid research. He was not interested in sexuality or in the controversy over innate versus acquired traits or of preformation versus epigenesis. He was interested only in verifying his hypothesis via the calculation of combinations 6. Mendel neglected everything that interested those who in reality were not his predecessors at all. The seventeenth-century ideology of hereditary transmission is replete with observations of animal and plant hybrids and monsters. Such curiosity served several purposes. It supported one side or the other in the debates between preformationists and epigenesists, ovists and animalculists. As a result, it was useful in resolving legal questions concerning the subordination of the sexes, paternity, purity of blood lines, and the legitimacy of the aristocracy. These concerns were not unrelated to the controversy between innatism and sensualism. The technology of hybridization was perfected by agronomists in search of advantageous varieties as well as by botanists interested in the relations between species. Only by isolating Maupertuis’s Vénus physique from its context can it be compared with the Versuche über Pflanzenhybriden. Mendel’s science is not the end point of a trail that can be traced back to the ideology it replaced, for the simple reason that that ideology followed not one but several trails, and none was a course set by science itself. All were rather legacies of various traditions, some old, others more recent. Ovism and animalculism were not of the same age as the empirical and mythological arguments advanced in favor of aristocracy. The ideology of heredity 7 was excessively and naively ambitious. It sought to resolve a number of important theoretical and practical legal problems without having examined their foundations. Here the ideology simply withered away by attrition. But the elimination of its scientific underpinnings brought it into focus as an ideology. The characterization of a certain set of observations and deductions as an ideology came after the disqualification of its claim to be a science; this was accomplished by the development of a new discourse, which circumscribed its field of validity and proved itself through the consistency of its results.

Instructive as it is to study the way in which scientific ideologies disappear, it is even more .instructive to study how they appear. Consider briefly the genesis of a nineteenth-century scientific ideology, evolutionism. The work of Herbert Spencer makes an interesting case study. Spencer believed that he could state a universally valid law of progress in terms of evolution from the simple to the complex through successive differentiations. Everything, in other words, evolves from more to less homogeneity and from lesser to greater individuation: the .solar system, the animal organism, living species, man, society, and · the products of human thought and activity, including language. Spencer explicitly states that he derived this law of evolution by· generalizing the principles of embryology contained in Karl-Ernst von Baer’s Uber Entwickelungsgeschichte der Thiere (1828). The publication of the Origin of Species in 1859 confirmed Spencer’s conviction that his generalized theory of evolution shared the scientific validity of Darwin’s biology. But he also claimed for his law of evolution the support of a science more firmly established than the new biology, claiming to have deduced the phenomenon of evolution from the law of conservation of energy, which he maintained could be used to prove that homogeneous states are unstable. If one follows the development of Spencer’s work, it seems clear that he used von Baer’s and, later, Darwin’s biology to lend scientific support to his views concerning social engineering in nineteenth-century English industrial society, in particular, his advocacy of free enterprise, political individualism, and competition. From the law of differentiation he deduced that the individual must be supported against the state. But perhaps this “deduction” was contained in the principles of the Spencerian system from the very beginning.

The laws of mechanics, embryology, and evolution cannot validly be extended beyond the domain proper to each of these sciences. To what end are specific theoretical conclusions severed from their premises and applied out of context to human experience in general, particularly social experience? To a practical end. Evolutionist ideology was used to justify industrial society as against traditional society on the one hand and the demands of workers on the other. It was in part antitheological, in part antisocialist. Thus evolutionist ideology was an ideology in the Marxist sense: a representation of nature or society whose truth lay not in what it said but in what it hid. Of course evolutionism was far broader than Spencer’s ideology. But Spencer’s views had a lasting influence on linguists and anthropologists. His ideology gave meaning to the word primitive and salved the conscience of colonialists. A remnant of its legacy can still be found in the behavior of advanced societies toward so-called underdeveloped countries, even though anthropology has long since recognized the plurality of cultures, presumably making it illegitimate for any one culture to set itself up as the yardstick by which all others are measured. In freeing themselves from their evolutionist origins, contemporary linguistics, ethnology, and sociology have shown that an ideology disappears when historical conditions cease to be compatible with its existence. The theory of evolution has changed since Darwin, but Darwinism is an integral part of the history of the science of evolution. By contrast, evolutionist ideology is merely an inoperative residue in the history of the human sciences.


With these examples I hope that I have clarified the way in which scientific ideologies come into being. Let me add that one must be careful not to confuse scientific ideologies with the ideologies of scientists, by which I mean ideologies that scientists engender when they attempt to systematize their research methods and procedures or when they talk about the place of science within the larger culture. The ideologies of scientists should perhaps be called ideologies of philosophers, that is, scientific-sounding doctrines propounded by men who in this realm are scientists only in a presumptive or presumptuous sense. In the eighteenth century the concepts of Nature and Experience were ideological concepts of scientists. By contrast, the concepts of “organic molecule” (Buffon) and “chain of being” (Bonnet) were concepts of scientific ideology in natural history.

To sum up:

a. Scientific ideologies are explanatory systems that stray beyond their own borrowed norms of scientificity.

b. In every domain scientific ideology precedes the institution of science. Similarly, every ideology is preceded by a science in an adjunct domain that falls obliquely within the ideology’s field of view.

c. Scientific ideology is not to be confused with false science, magic, or religion. Like them, it derives its impetus from an unconscious need for direct access to the totality of being, but it is a belief that squints at an already instituted science whose prestige it recognizes and whose style it seeks to imitate.

Let me conclude by going back to where I began to propose a theory that .may shed some light on the practice of the history of science. A history of science that views science as a series of articulated truths need not concern itself with ideology. Historians of science who hold this view naturally leave questions of ideology to the historians of ideas or, worse still, the philosophers.

A history of science that views science as a progressive process of purification governed by norms of verification cannot fail to concern itself with scientific ideology. Gaston Bachelard distinguished between obsolete and valid science, and while it is wise to separate one from the other, it is also wise to study how the two are related. The obsolete is condemned in the name of truth and objectivity. But what is now obsolete was once considered objectively true. Truth must submit itself to criticism and possible refutation or there is no science.

Distinguishing between ideology and science prevents us from seeing continuities where in fact there are only elements of ideology preserved in .a science that has supplanted all earlier ideology. Hence such a distinction prevents us from seeing anticipations of the Origin of Species in Rousseau’s Dream of d’Alembert. Conversely, recognizing the connections between ideology and science should prevent us from reducing the history of science to a featureless landscape, a map without relief.

The historian of science must work and must present his work on two levels. If he fails to recognize and incorporate scientific ideology into his work, he runs the risk of producing nothing more than ideology himself, by which I mean in this instance a history that is a false consciousness of its object. The closer the historian thinks he comes to his object, the farther he is from the target. His knowledge is false knowledge, because true critical knowledge requires critical perspective; the historian cannot accurately see any object that he does not actively construct. Ideology is mistaken belief in being close to truth. Critical knowledge knows that it stands at a distance from an operationally constructed object. Professor Suchodolski is right on one point: the history of truth is a contradiction in terms.

Georges Canguilhem.

Lecture given in October 1969 at Warsaw and Cracow to the Institute for the History of Science and Technology and the Polish Academy of Science. It was published in the journal Organon #7 (1970).

English translation by Arthur Goldhammer,
Massachusetts Institute of Technology, USA, 1988.


1 See Voltaire, “Prejudice”, in the Philosophical Dictionary.

2 Report entitled “Factors Aff ecting the Development of the History of Science”, XIIe Congres International d’Histoire des Sciences: Colloques, textes des rapports (Paris: 1968), p. 34.

3 “The contempt that [Napoleon] professed for industrial men of affairs complemented his contempt for the ideologues” (Marx, The Holy Family, 6.3.c).

4 In Marx’s view, the political ideologies of the French and English in the eighteenth century were less remote from their true base than the religious ideology of the Germans.

5 In the Communist Manifesto the illusion of the bourgoisie that the social situation that makes it the dominant class is eternal ii characterized as a “self-interested conception”.

6 Jacques Piquemal, “Aspects de la pensée de Mendel”, lecture delivered at the Palais de la Découverte in Paris, 1965.

7 In this case the name of the science was transferred post hoc to the ideology; in the case of atomism it was the other way around.

Votre commentaire

Entrez vos coordonnées ci-dessous ou cliquez sur une icône pour vous connecter:


Vous commentez à l’aide de votre compte Déconnexion /  Changer )

Photo Google

Vous commentez à l’aide de votre compte Google. Déconnexion /  Changer )

Image Twitter

Vous commentez à l’aide de votre compte Twitter. Déconnexion /  Changer )

Photo Facebook

Vous commentez à l’aide de votre compte Facebook. Déconnexion /  Changer )

Connexion à %s